Patients with rapid-cycling bipolar disorder (RCBD) experience at least four episodes of affective illness (depression, hypomania, and/or mania) in a year. Their mood cycles are associated with changes in the sleep- wake cycle, and sleep deprivation experiments demonstrate that these sleep-wake changes are not merely symptoms of the illness, but have pathogenic significance. These observations raise the question of whether circadian dysregulation and/or dysregulation in the sleep-wake cycle are involved in the pathogenesis of RCBD, and whether stabilization of these physiological systems could have beneficial clinical effects. In this project we are performing a series of cross-sectional and longitudinal studies designed to address the first of these questions. In the cross-sectional study, patients are admitted to the hospital for brief, intensive evaluations of circadian rhythms and sleep once shortly after they switch into hypomania, and once shortly after they switch into depression. The hospitalizations are each 48 hours in duration and consist of a 20-hour "naturalistic day," during which patients sleep and eat ad lib, and a 28-hour constant routine designed to minimize the effects of light, sleep, activity, and caloric loading on the patient's circadian rhythms. During the hospitalizations, patients wear rectal temperature probes, sleep is monitored by EEG, and blood is withdrawn every thirty minutes through an indwelling intravenous catheter. Blood samples are used to measure melatonin, TSH, cortisol, prolactin and growth hormone. Data from the first twelve patients indicate that there is a trend for the onset of melatonin secretion to be shifted earlier in hypomania, compared with depression. The other hormonal data are currently being analyzed. Analysis of the EEG data does not show any differences between the two mood states. Data from a small number of patients studied longitudinally are consistent with these cross-sectional data in generally showing a delay is maximal (i.e. early vs. late in the episode). In addition to these studies, we have used data from controls participating in other studies in the Branch to obtain preliminary information as to whether, and how, patients' and controls' sleep and melatonin rhythms may differ. This analysis shows that patients with RCBD have relatively delayed time of nocturnal melatonin onset and of sleep onset. In addition, patients with RCBD appear to have decreased duration of melatonin secretion. Based on these data, we propose a model whereby the symptoms of the illness (fluctuating levels of activation and fluctuating sleep duration) cause unstable entrainment, which in turn exacerbates the instability in the sleep-wake cycle and the symptoms of the illness. According to this model, stabilizing circadian rhythms could have beneficial clinical effects, a hypothesis that we test in ZO1 MH 02688-03 CPB.